Prevention of gram-negative bacillary pneumonia using aerosol polymyxin as prophylaxis. I. Effect on the colonization pattern of the upper respiratory tract of seriously ill patients.

A prospective study used polymyxin B by aerosol to reduce colonization of the upper respiratory tract with nosocomial gram-negative bacilli. 58 high-risk patients from the Respiratory-Surgical Intensive Care Unit entered the trial. 33 were randomly selected to receive 2.5 mg/kg/day of polymyxin B by hand atomizer into the pharynx, and tracheal tube if present. 17 of 25 control patients became colonized with gram-negative bacilli as compared with 7 of 33 polymyxin-treated patients (p < 0.01). Control patients became colonized with a total of 33 gram-negative bacilli: 3 were Pseudomonas aeruginosa, 21 were species of Enterobacteriaceae. The polymyxin-treated patients became colonized with a total of 11 gram-negative bacilli: no P. aeruginosa and only 3 species of Enterobacteriaceae were recovered. Colonization increased with duration in Respiratory-Surgical Intensive Care Unit and with time of required controlled ventilation. Polymyxin most effectively prevented the increase in colonization in treated patients who stayed in the Respiratory-Surgical Intensive Care Unit for longer than 1 wk and who required controlled ventilation for at least 72 h.

Prevention of gram-negative bacillary pneumonia using polymyxin aerosol as prophylaxis. II. Effect on the incidence of pneumonia in seriously ill patients.

All 744 patients admitted to a Respiratory-Surgical Intensive Care Unit (RSICU) were included in a prospective study of the effects of a polymyxin (2.5 mg/kg body wt/day in six divided doses) or a placebo aerosol sprayed into the posterior pharynx and tracheal tube (if present), during 11 alternating 2-mo treatment cycles. The incidence of upper airway colonization in the RSICU with Pseudomonas aeruginosa was 1.6% during the polymyxin treatment cycles (total 374 patients) and 9.7% during the placebo cycles (370 patients) (X2 equals 23.2, P less than 0.01). 3 patients in the RSICU acquired Pseudomonas pneumonia, as defined by independent "blinded" assessors, during the polymyxin cycles while 17 acquired a Pseudomonas pneumonia during the placebo cycles (X2 equals 10.2, P less than 0.01). The overall mortality was similar in both placebo and polymyxin-treated groups (12.2 vs. 12.0%). Systemic antibiotic usage was similar in the different cycles; 49% of patients in the placebo and 53% in the polymyxin-treated groups received systemic antibiotics while in the RSICU.

Resource utilization among intensive care patients. Managed care vs traditional insurance.

BACKGROUND: There is considerable evidence that members of managed care organizations use fewer hospital resources than patients covered by traditional health insurance. While intensive care might seem to be an unlikely setting for such differences to exist, the relationship between health coverage and use of intensive care has not been examined. METHODS: We conducted a cross-sectional analysis of consecutive intensive care unit admissions at a regional tertiary care teaching hospital. Patients in managed care plans (n = 159) and with traditional insurance (n = 389) were compared with respect to length of stay, hospital charges, charges for specific services, and use of mechanical ventilation. The analysis controlled for severity of illness, as measured by the Mortality Probability Model, case mix, and mortality. The whole sample as well as subsamples representing medical, emergency surgery, and elective surgery patients were examined. RESULTS: The managed care group, on average, had short stays (both hospital and intensive care unit), lower charges, and less use of mechanical ventilation than the traditionally insured group. Average differences of about 30% to 40% were observed. The finding held for the whole sample as well as the medical and emergency surgery subsamples. The differences were more pronounced in the patients with lowest severity of illness. CONCLUSION: Even in a setting where there would appear to be relatively little room for discretion in treatment decisions, incentives associated with type of health insurance seemed to affect resource use.

The relationship between age and the use of DNR orders in critical care patients. Evidence for age discrimination.

OBJECTIVE: To determine whether a relationship exists between the use of do-not-resuscitate (DNR) orders in the intensive care unit and the age of the patient after controlling for the severity of illness. METHODS: Patients from the Mortality Probability Model database, which includes 6103 patients in 4 large hospitals, and from a second database, which includes 3226 additional patients in 25 hospitals, were analyzed through logistic regression seeking a relationship between age and DNR use. Adult medical and surgical intensive care units from 27 hospitals in the United States were included. RESULTS: In the Mortality Probability Model database, 11.4% of the patients had DNR orders written. In the group of patients younger than 65 years, 8% had DNR orders. This percentage climbed rapidly with age. For age ranges of 65 to younger than 75 years, 75 to younger than 85 years, and 85 years or older, the percentage of patients who had a DNR order was 11.2%, 18.9%, and 32.6%, respectively. Similar results were found in the second database: 5.4% of patients had DNR orders and, again, the rise in the use of DNR orders was associated with increased age. For patients younger than 75 years, 4.2% had DNR orders. For the older groups, 75 to younger than 85 years and 85 years and older, the rates were 8.8% and 15.4%, respectively. Logistic regression was used to control for severity of illness; when compared with patients younger than 65 years, patients 75 to younger than 85 years were 50% more likely to have DNR orders written and patients 85 years or older were 140% more likely to have DNR orders written. CONCLUSIONS: Older patients (> or = 75 years old) are significantly more likely than younger patients to have DNR orders written even after the severity of illness is controlled as a confounding variable. This association suggests age discrimination and becomes stronger as patient age increases.

Enhanced bleeding with cefoxitin or moxalactam. Statistical analysis within a defined population of 1493 patients.

Most cases of beta-lactam-associated coagulopathy occur in patients with other risk factors. This study analyzed temporally related clinical bleeding events in 1493 patients who received one antibiotic for at least three days. Univariate and multivariate analyses controlled for condition variables (nutritional status, renal, hepatic, or hematologic dysfunction, intensive care unit stay) and treatment variables (use of antiplatelet agents, anticoagulants, vitamin K, antitumor chemotherapy or antiulcer therapy, steroids) that could have been associated with bleeding independently. Rates of bleeding ranged from 0% (chloramphenicol sodium succinate, vancomycin hydrochloride, erythromycin lactobionate) to 8.2% (cefoxitin) to 22.2% (moxalactam disodium). Multiple logistic regression analyses revealed that only moxalactam (odds ratio, 9.9) and cefoxitin (odds ratio, 2.1) exhibited significantly higher likelihoods of bleeding than other agents. This study statistically confirms increased risk of bleeding with moxalactam, heretofore reported only anecdotally. Cefoxitin may carry risks greater than previously believed.

Intensive care unit patient follow-up. Mortality, functional status, and return to work at six months.

Six months after hospital discharge, we followed up 1545 patients who had received care in the general medical-surgical intensive care unit (ICU) of a tertiary care hospital. Vital status could not be ascertained for 200 of these patients. Of the 1345 former ICU patients for whom a determination of vital status could be made, 1261 (94%) were alive and 84 (6%) had died. Of those known to be living, 887 (70%) responded to a questionnaire regarding employment, functional, and social status. A large proportion of survivors less than 40 years of age had returned to work. Younger patients admitted to the hospital for elective surgery reported as much compromise of physical and psychological activity as did older patients admitted for emergency reasons. Older survivors reported an increase of interaction with family members and a decrease of social interaction with those other than family.

Therapeutic impact of pulmonary artery catheterization in a medical/surgical ICU.

The objective of this study was to determine the following: (1) if standard clinical evaluation is sufficient to provide an accurate estimate of hemodynamic status of unstable ICU patients; (2) the impact of pulmonary artery catheterization (PAC) on diagnosis and treatment plan; and (3) whether therapy provided after PAC was appropriate as judged by an expert panel of senior ICU physicians. A descriptive analysis of utilization of pulmonary artery catheters in a medical/surgical ICU population was performed in a university-affiliated hospital (24-bed medical/surgical ICU). The subjects included 154 medical/surgical patients judged by ICU residents and attendings to require PAC. All 154 patients underwent PAC with four patients having more than one catheterization. Prior to insertion of the catheter, a questionnaire was completed by medical/surgical residents and attendings indicating reasons for PAC insertion and estimate of hemodynamics. Following PAC, residents/attendings indicated their evaluation of hemodynamics and planned therapy. An expert panel rated performance of the house staff regarding treatment plan on a scale of 1 to 5 (5 indicating optimal therapy). The overall proportion correct classification for pulmonary artery wedge pressure (PAWP), CO, and systemic vascular resistance (SVR) were 47 percent, 51 percent, and 36 percent, respectively. In 45 percent of PAC, information obtained resulted in a major change in therapy. Major change in therapy occurred more often when prediction of PAWP by residents proved inaccurate. The expert panel judged appropriate scores of 3, 4 and 5 in 84 percent of the cases. Prediction of hemodynamics in ICU patients by clinical evaluation alone is inaccurate and unreliable. There is a positive correlation between inaccurate prediction of hemodynamics and major therapeutic changes after PAC. Most resident/attending performance was judged appropriate. Results of this study suggest that PAC was instrumental to the management scheme in many patients unresponsive to initial therapy. However, a subset of ICU patients were judged to have been managed favorably, yet had treatment based on inaccurate hemodynamic assessment.

Outbreak of nosocomial Flavobacterium meningosepticum respiratory infections associated with use of aerosolized polymyxin B.

Flavobacterium meningosepticum is an uncommon cause of adult nosocomial infection. On a medical/surgical intensive care unit we recently encountered an adult outbreak of respiratory colonization and infection caused by this organism, which was associated with the prophylactic use of aerosolized polymyxin B that had been used in an attempt to abort an outbreak of infection caused by highly resistant strains of Pseudomonas aeruginosa. Twenty isolates (95% from respiratory secretions) of F. meningosepticum from nine persons were identified during a 2 1/2-month period. No environmental source has been identified to date. Pneumonia developed in five patients, and two deaths associated with this organism occurred. All isolates were sensitive to ciprofloxacin; none were sensitive to other antibiotics tested, including third-generation cephalosporins, aminoglycosides, erythromycin, trimethoprim-sulfamethoxazole, antipseudomonal penicillins, aztreonam, and imipenem/cilastatin. Two patients with nosocomial pneumonia were successfully treated with oral ciprofloxacin. F. meningosepticum may emerge as an important pathogen if prophylactic use of polymyxin B becomes more widespread. Ciprofloxacin may become the agent of choice for treatment of this organism.

Trends from the United States with end of life decisions in the intensive care unit.

OBJECTIVE: To describe the changes that have occurred in the United States since medicine has moved away from a paternalistic model to one that promotes patient autonomy and self-determination. To discuss the implications for cardiopulmonary resuscitation (CPR) and the increasing use of when not to perform CPR and other life-sustaining therapies. To describe the various interpretations of the ritual term Do-Not-Resuscitate (DNR) and to introduce the concept of futility in the context of non-beneficial over-treatment and discriminatory under-treatment. SETTING: Selected clinical, philosophical and public policy literature and two illustrative case examples. RESULTS: 1. There is no longer a mandate to perform CPR on all dying patients, even though the Council on Ethical and Judicial Affairs of the American Medical Association in 1991 said that the only restrictions should be in patients with an irreversible terminal condition or when the physician writes the order, DNR. 2. The DNR order usually requires the informed refusal of CPR by the patient or family. There is only minimal support for a unilateral decision even for patients with far advanced disease. 3. DNR is often the first step in the negotiated process of forgoing care in the ICU. There are multiple interpretations of DNR both in and outside of the ICU. 4. Health Proxy is the latest attempt to have a person clarify his/her wishes and preferences by naming a decision maker, if the individual losses mental capacity. 5. Although ethical principles seem well established, there are inconsistent interpretations and practices at the bedside in the United States in part due to the restructuring of the relationship between physicians and patients, providers and consumers/clients. 6. Objective severity scores such as Apache III, SAPS II, MPM II are generally not applicable for individual patient end-of-life decisions. CONCLUSIONS: Although Health Proxy in its current formulation has been disappointing, there is a clear trend for wider application of DNR and for more active discussions about withholding or forgoing other life-sustaining therapies. DNR has a different interpretation late into the ICU course (> 72 h) than when applied at or shortly after ICU admission. Late in the ICU course, it has been decided by the medical team and family or surrogate decision maker/Health Proxy that the patient has failed or is in the process of failing aggressive ICU therapy. Early use of DNR may be related to limitations based on pre-existing chronic or subacute disease burden or an unwillingness to proceed with a full ICU course of therapy. It is unclear how Ethics Committees, risk management and hospital administrators, national practice guidelines, governmental sponsored health care reform will interface with the highly complex individual patient--physician--family--Health Proxy interface as practiced in the United States. Dialogue between the Society of Critical Care Medicine and the European Society of Critical Care Medicine and among interested physicians could provide a format for a multi-cultural context to discuss end of life issues in the ICU setting.

Outcome prediction for individual intensive care patients: useful, misused, or abused?

Probabilities of hospital mortality provide meaningful information in many contexts, such as in discussions of patient prognosis by intensive care physicians, in patient stratification for analysis of clinical trial data by researchers, and in hospital reimbursement analysis by insurers. Use of probabilities as binary predictors based on a cut point can be misleading for making treatment decisions for individual patients, however, even when model performance is good overall. Alternative models for estimating severity of illness in intensive care unit (ICU) patients, while demonstrating good agreement for describing patients in the aggregate, are shown to differ considerably for individual patients. This suggests that identifying patients unlikely to benefit from ICU care by using models must be approached with considerable caution.

Propofol-containing sedatives increase levels of parathyroid hormone.

OBJECTIVE: To evaluate the effects of propofol and propofol containing disodium edetate (ethylenediaminetetraacetic acid [EDTA]) on the parathyroid-calcium axis in normal subjects. DESIGN: Randomised, double-blind, age-stratified, crossover trial. SETTING: Single centre. PATIENTS: A total of 50 healthy subjects. INTERVENTIONS: Each subject was randomised to receive propofol or propofol containing EDTA on day 1 and the alternate treatment between days 15 and 29, with a 2-week wash-out period in between. On the day of treatment, subjects received a bolus of trial medication (1 or 2 mg/kg) followed by a 60-minute observation period. At the end of 60 minutes, subjects received trial medication infused for 60 minutes at 1 of 4 randomised infusion rates (25, 50, 100, or 200 microg/kg per min). Subjects were monitored for an additional 60 minutes following the infusion. MEASUREMENTS AND RESULTS: Blood pressure, heart rate, respiratory rate, oxygen saturation, blood ionised calcium concentration, serum total magnesium concentration, serum intact parathyroid hormone (PTH) level, and plasma EDTA level were assessed at periodic intervals during and following the bolus and continuous infusion of trial medication. Mean arterial pressure significantly decreased (p < 0.05) following the bolus injection of both trial medications and returned to baseline at 60 minutes; it significantly decreased again during the continuous infusion and returned to baseline during recovery. Heart rate and respiratory rate fluctuated in both groups with significant increases and decreases throughout the study period following the bolus injection; both returned to baseline during the recovery period in each group. Ionised calcium and total magnesium concentrations remained within normal limits and were unchanged in response to both study medications. PTH levels significantly increased following the bolus injection of both study drugs. The increase in PTH levels was greater with higher doses of study medication during the infusion period. There was no difference in the response of blood pressure, heart rate, respiratory rate, or PTH levels between propofol and propofol with EDTA. EDTA levels increased significantly during the infusion of propofol with EDTA, reaching mean levels of 240 ng/mL. CONCLUSIONS: The results of this study indicate that propofol increases PTH levels in normal subjects; however, propofol with EDTA does not alter ionised calcium or total magnesium concentrations.

Cation metabolism during propofol sedation with and without EDTA in patients with impaired renal function.

OBJECTIVE: To compare the effects of propofol with and without disodium edetate (EDTA) on cation metabolism in intensive care unit (ICU) patients with renal insufficiency who received propofol or propofol plus EDTA (propofol EDTA) for sedation and mechanical ventilation. DESIGN: Double-blind, randomised, multicentre study. SETTING: Medical and surgical ICUs from 5 hospitals. PATIENTS: Thirty-nine ICU patients with acute and chronic renal impairment expected to require at least 24 hours of continuous sedation and respiratory failure necessitating mechanical ventilation. INTERVENTIONS: Propofol or propofol EDTA administered for sedation by continuous intravenous infusion. MEASUREMENTS AND RESULTS: The depth of sedation, as measured by the Modified Ramsay Sedation Scale, was similar in the 2 groups, when adjusted for dosing differences. The amount of propofol required to maintain adequate sedation was decreased in both groups compared to propofol requirements in ICU patients with normal renal function. EDTA levels were elevated at baseline in both groups. In the propofol EDTA group, the EDTA levels increased further by 20 % but decreased to below baseline EDTA levels at 48 hours after sedation. In the propofol group, EDTA levels decreased during sedation and remained below baseline levels at 48 hours after sedation. PATIENTS in both groups were hypocalcaemic and hyperphosphataemic at baseline with low levels of 1,25-dihydroxyvitamin D and elevated parathyroid hormone (PTH) levels. Other than a slight difference in ionised serum calcium levels at 4 h after the start of sedation, there were no significant differences observed in serum calcium levels between the two groups. There were no significant differences in 1,25-dihydroxyvitamin D or PTH levels over time between the two groups. There was no significant effect on renal function in either group. CONCLUSIONS: The results of this study suggest that adding EDTA to propofol does not adversely affect cation homeostasis or renal function when used for sedation of ICU patients with renal insufficiency. Although EDTA levels increased over time from baseline levels in patients with renal insufficiency who receive propofol EDTA, this increase does not appear to be clinically significant, and EDTA levels return to below baseline levels within 48 hours of discontinuing the propofol EDTA infusion. The efficacy of propofol with and without EDTA also appears comparable in these patients.

Safety and efficacy of propofol with EDTA when used for sedation of surgical intensive care unit patients.

OBJECTIVE: To compare propofol with disodium edetate (EDTA) and propofol without EDTAwhen used for the sedation of critically ill surgical intensive care unit (ICU) patients. DESIGN: Prospective, randomised, multicentre trial. PATIENTS: A total of 122 surgical ICU patients who required intubation and mechanical ventilation. INTERVENTIONS: Patients were randomised to receive either propofol or propofol plus EDTA (propofol EDTA) by continuous infusion for sedation. MEASUREMENTS AND RESULTS: The addition of EDTA to propofol had no effect on calcium or magnesium homeostasis, renal function, haemodynamic function, or efficacy when used for the sedation of surgical patients in the ICU. The most common adverse events were hypotension, atrial fibrillation, and hypocalcaemia. In this trial, a greater number of serious adverse events and adverse events leading to withdrawal occurred in the propofol group relative to the propofol EDTA group. There was a significantly lower crude mortality rate at 7 and 28 days for the propofol EDTA group compared with the propofol group. There were no statistically significant differences between groups with respect to depth of sedation. CONCLUSION: The propofol EDTA formulation had no effect on calcium or magnesium homeostasis, renal function, or sedation efficacy compared with propofol alone when used for sedation in critically ill surgical ICU patients. There was a significant decrease in mortality in the propofol EDTA group compared with the propofol group. Further investigations are needed to validate this survival benefit and elucidate a possible mechanism.

Trace element homeostasis during continuous sedation with propofol containing EDTA versus other sedatives in critically ill patients.

OBJECTIVE: To evaluate changes in serum and urinary zinc, cobalt, copper, iron, and calcium concentrations in critically ill patients receiving propofol containing disodium edetate (disodium ethylenediaminetetraacetic acid [EDTA]) versus sedative agents without EDTA. DESIGN: This was a randomised, open-label, parallel-group study with randomisation stratified by baseline Acute Physiology and Chronic Health Evaluation (APACHE II) scores. SETTING: Intensive care units (ICU) in 23 medical centres. PATIENTS: Medical, surgical, or trauma ICU patients 17 years of age or older who required mechanical ventilator support and sedation. INTERVENTIONS: A total of 106 patients received propofol containing 0.005 % EDTA (propofol EDTA), and 104 received other sedative agents without EDTA (non-EDTA). Only the first 108 patients were assessed for urinary trace metal excretion. Twenty-four-hour urine samples were collected on days 2, 3, and 7 and every 7 days thereafter for determination of zinc, cobalt, copper, iron, and calcium excretion; EDTA levels; urine osmolality; albumin levels; and glucose levels. The first 143 patients were assessed for serum concentration of zinc, cobalt, copper, iron, and calcium; creatinine; blood urea nitrogen; and albumin at baseline and once during each 24-hour urine collection. MEASUREMENTS AND RESULTS: For the assessment of trace metals, patients receiving propofol EDTA demonstrated increased mean urinary excretion of zinc, copper, and iron compared with the normal range. All patients receiving sedatives demonstrated increased urinary excretion of zinc and copper above normal reference values. Compared with the non-EDTA sedative group, the propofol EDTA group demonstrated increased urinary excretion of zinc and iron. Mean serum concentrations of zinc and total calcium were decreased in both patient groups. Serum zinc concentrations increased from baseline to day 3 in the non-EDTA sedative group but not in the propofol EDTA group. Renal function, measured by blood urea nitrogen, serum creatinine, and creatinine clearance, did not deteriorate during ICU sedation with either regimen. CONCLUSION: This study showed that critical illness is associated with increased urinary losses of zinc, copper, and iron. Propofol EDTA-treated patients had greater urinary losses of zinc and iron and lower serum zinc concentrations compared with the non-EDTA sedative group. No adverse events indicative of trace metal deficiency were observed in either group. The clinical significance of trace metal losses during critical illness is unclear and requires further study.

Nosocomial pneumonia during stress ulcer prophylaxis with cimetidine and sucralfate.

BACKGROUND: Recent studies have questioned the use of histamine (H2) receptor antagonist in stress ulcer prophylaxis because of an increased incidence of nosocomial pneumonia and subsequent death. DESIGN: This prospective randomized study compared prophylaxis with cimetidine vs sucralfate. SETTING: Medical/surgical intensive care unit in Springfield, Mass. PATIENTS: One hundred fourteen patients were enrolled. INTERVENTIONS: Cimetidine, administered as a primed continuous infusion using a 300-mg bolus followed by 37.5 mg/h, was compared with sucralfate, administered via nasogastric tube, at a dosage of 1 g every 6 hours suspended in 20 mL of sterile water. MAIN OUTCOME MEASURES: End points of the study included nosocomial pneumonia, gastrointestinal hemorrhage, and death. RESULTS: Fifty-six patients were randomized to receive cimetidine and their rate of pneumonia was 12.5%; upper gastrointestinal hemorrhage, 3.6%; and mortality, 33.9%. Fifty-eight patients were given sucralfate, and their rate of pneumonia was 13.8%; upper gastrointestinal hemorrhage, 3.4%; and mortality, 37.9%. There were no significant differences between these study end points. In patients who had pneumonia, 80% of isolates were aerobic gram-negative bacilli. CONCLUSIONS: These observations suggest that the rate of nosocomial pneumonia is not increased in patients in the intensive care unit who receive prophylaxis with cimetidine to prevent stress ulcer bleeding.

Clinical assessment of hemodynamic values in two surgical intensive care units. Effects on therapy.

A prospective study of 126 surgical patients from two institutions was undertaken to assess the impact of pulmonary artery catheterization in surgical intensive care units. Before catheterization, surgical residents were asked to predict pulmonary artery wedge pressure, cardiac output, systemic vascular resistance, and plan of therapy. After catheterization, each chart was reviewed by a panel of intensive care specialists and a general surgeon. Correct classification for the hemodynamic variables ranged from 47% to 55%. Catheterization results prompted a major change in therapy in 50% of patients. The data suggest that hemodynamic variables obtained from pulmonary artery catheterization improve the accuracy of bedside evaluation and lead to alteration in therapy, particularly in patients whose pulmonary artery wedge pressure predictions were poor.

Synergistic effect of acute renal failure and respiratory failure in the surgical intensive care unit.

A retrospective evaluation of the effect of renal and respiratory failure on mortality in our surgical intensive care unit was undertaken. The coexistence of combined renal and respiratory failure had a synergistic adverse effect on survival. Combined pulmonary and kidney failure appeared to develop simultaneously. A subset of patients with severe prerenal azotemia but without uremia had the highest mortality. These results are not consistent with the simple combination of single systems failure but rather suggest that renal and respiratory failure are makers of a generalized underlying defect.

Using severity measures to describe high performance intensive care units.

This article describes the use of various scores and probabilities to clinically categorize patients in the adult intensive care unit. Some of the limitations of these severity measures are reviewed including variable definitions, timing of measurements, and whether models can be used for individual patients. Also, this article discusses how probability models may be used to compare similar types of intensive care units using standardized clinical and cost performance indices.

Why severity models should be used with caution.

There are now two validated time points for predicting hospital mortality of ICU patients--at admission and at 24 hours. The best purposes include evaluation of high clinical performance ICUs and for patients being enrolled in clinical trials. For the latter purpose, the model must be calibrated in the individual hospital to ensure that the model is applicable. This can be estimated by using goodness-of-fit testing. There are fewer uses for physiology scores and increased emphasis on converting scores to probabilities. For individual patient application, the model should be demonstrated to have high discrimination, as measured by the area under the receiver operating characteristic curve, and high calibration, as defined by goodness-of-fit testing. Although models have improved substantially and are now based on much larger databases, there is considerable uncertainty in their application for insurance purposes, triage, regulatory applications, sanctions against individual physicians, and cost containment. Current models may not adequately describe important ICU conditions such as adult respiratory distress syndrome and multi-organ dysfunction occurring after 24 hours into ICU care. For family discussions regarding prognosis of individual patients, ICU severity models must be used cautiously at admission or after 24 hours, with the understanding of the strengths and weakness of estimating probabilities of hospital mortality. The mathematical link between physiology score and estimation of hospital mortality is established only for the time point of 24 hours after ICU admission. Calibration and discrimination of the admission and 24-hour models also must be performed within each hospital in which individual probabilities are presented to families. It may be possible to customize a probability model such as MPM to achieve a high level of calibration at the individual hospital level.

Lymphocytopenia in the surgical intensive care unit patient.

Total lymphocyte counts were monitored in 328 consecutive adult patients in the intensive care unit of a large community hospital. Lymphocytopenia was common with a severe reduction in circulating lymphocytes (less than 900/mm3) present in 37% of the patients and moderate reduction (900-1500/mm3) in 38% of the patients. Mortality was 30% and 13% in these two groups, respectively. Patients undergoing high-risk elective surgery frequently had moderate lymphocytopenia even when corticosteroids were not administered. Patients having two or more clinically evident infections with severe lymphocytopenia had a 59% mortality rate. Among patients who did not manifest any clinically detectable infection, mortality associated with those having severe lymphocyte reduction was significantly greater than those with normal lymphocyte counts or moderate depletion.